CERNbox: experiences with ownCloud service deployment

The CERNbox pilot was started in March 2014 as a first step to introducing a file sync and share service at CERN. CERNbox is a future extension of our current data storage services portfolio for the CERN user community (100PB total storage capacity). We present our experience in deploying the pilot service, user feedback and operational challenges as well as the future evolution of the service towards a high-scalability, heavy-duty service for handling datasets at a large scale. Firstly, we want to make file syncing and sharing as easy and accessible as it is with popular cloud services offered by external providers such as Dropbox or OneDrive. The ownCloud software stack does the job: it provides the ease-of-use, integration in the desktop and mobile environments and user-friendly interfaces which are already good enough to compete with commercial alternatives. Moreover, ownCloud gets even better at doing this with every release. Secondly, we want the CERNbox service to take full advantage of existing Petabyte-range data stores at CERN and achieve corresponding scalability of the sync layer of the service. This introduces new architectural challenges but also opens up an evolutionary path for advanced use-cases, well beyond the concept of low-frequency synchronization of a relatively modest number of document files.



  • Jakub T. Moscicki
  • M. Lamanna

Part of session

Emerging cloud stack

Related documents